Intellectual Property in Contracts

Before you can evaluate any IP clause, you need to understand what types of intellectual property exist — because each type carries different ownership rules, registration requirements, protection periods, and transfer mechanics. A single contract may involve all four categories simultaneously, and the legal rules governing each are different.

Copyright (17 U.S.C. §§ 101–1332). Copyright protects original works of authorship fixed in a tangible medium — including software code, written content, graphic designs, photographs, videos, architectural drawings, and musical compositions. Copyright arises automatically upon creation; no registration is required for protection to exist, although registration with the U.S. Copyright Office is required before filing an infringement lawsuit and provides access to statutory damages ($750–$150,000 per infringed work) and attorney's fee awards. Copyright protection lasts for the life of the author plus 70 years (or, for works made for hire, 95 years from first publication or 120 years from creation, whichever is shorter). In contracts, copyright ownership is the most commonly disputed IP category — especially between businesses and independent contractors.

Patents (35 U.S.C. §§ 1–390). Patents protect novel, non-obvious, and useful inventions — including processes, machines, manufactures, compositions of matter, and (in the U.S.) software methods and business processes that satisfy the subject-matter eligibility requirements of 35 U.S.C. § 101 as interpreted after Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014). A utility patent provides protection for 20 years from the filing date (subject to maintenance fee payments). Patent rights must be actively claimed through prosecution before the USPTO; unlike copyright, no patent protection arises automatically. In employment and contractor agreements, invention assignment clauses transfer ownership of patentable inventions from the creator to the contracting party.

Trademarks (15 U.S.C. §§ 1051–1141n — Lanham Act). Trademarks protect brand identifiers — words, logos, slogans, colors, sounds, and trade dress that distinguish the source of goods or services in commerce. Trademark rights arise from use in commerce, not from registration, but federal registration with the USPTO provides nationwide priority, the right to use the ® symbol, and access to Customs recordation to block infringing imports. In contracts, trademark licenses are common in franchise agreements, brand partnership agreements, and distribution agreements. The trademark licensor must maintain quality control over licensed goods and services to preserve the trademark's validity — failure to do so (a "naked license") can result in trademark abandonment.

Trade Secrets (18 U.S.C. § 1836 — Defend Trade Secrets Act; state law). Trade secrets are any information that: (1) derives independent economic value from not being generally known or readily ascertainable; (2) is subject to reasonable measures to maintain its secrecy. Trade secrets can include formulas, processes, algorithms, customer lists, pricing data, business methods, source code, and marketing strategies. Unlike patents, trade secret protection is theoretically unlimited in duration — as long as the information remains secret. The Defend Trade Secrets Act of 2016 (DTSA) provides a federal civil cause of action for trade secret misappropriation with damages including actual loss, unjust enrichment, and exemplary damages (up to 2x actual damages) for willful misappropriation, plus attorney's fees. Almost every employment and consulting agreement includes trade secret provisions — but the contractual provisions must be understood alongside state law, which often provides additional or different protections.

Why the Mix Matters. A software development contract, for example, may involve all four types simultaneously: copyrighted source code, patentable algorithms, trademarked UI elements, and trade secret architecture decisions. The ownership rules differ for each, and a contract that comprehensively assigns copyright may leave patent rights unaddressed — an oversight that creates co-ownership disputes when the contractor later files a patent application on the same invention.

The work-for-hire doctrine is the most significant copyright rule in any services contract. Understanding it — precisely — is essential because it determines who owns creative work without any need for a separate transfer, and because the rule operates very differently depending on whether the creator is an employee or an independent contractor.

Work Made for Hire: Two Paths. Under 17 U.S.C. § 101, a "work made for hire" is: (1) a work prepared by an employee within the scope of employment; OR (2) a work specially ordered or commissioned for use as one of nine specified categories, if the parties expressly agree in a written instrument signed by them that the work shall be considered a work made for hire.

Path 1: Employee Within Scope of Employment. If a W-2 employee creates a work within the scope of their employment duties, the employer is automatically the "author" and owner of that copyright — no written agreement is required. "Scope of employment" is determined by the agency test factors from Community for Creative Non-Violence v. Reid, 490 U.S. 730 (1989): the hiring party's control over the work, the skill required, the provision of tools and location, duration of the relationship, whether the work is within the regular business of the hiring party, and whether the parties treat themselves as employer/employee for tax and benefits purposes. A software engineer who writes code during business hours using company equipment for a product within the company's business line has almost certainly created a work made for hire. A software engineer who writes an independent app on personal time, using personal equipment, for a purpose unrelated to the employer's business, has almost certainly not.

Path 2: Independent Contractor — The 9 Statutory Categories. For independent contractors, work-for-hire status requires BOTH a written agreement AND the work must fall within one of these nine statutory categories: (1) contribution to a collective work; (2) part of a motion picture or other audiovisual work; (3) translation; (4) supplementary work (foreword, afterword, editorial notes, bibliography); (5) compilation; (6) instructional text; (7) test; (8) answer material for a test; (9) atlas. If the commissioned work does not fall within one of these nine categories — for example, a custom logo, website, standalone software application, or custom-written marketing copy — it CANNOT be a work made for hire for an independent contractor, regardless of what the contract says. The clause in the contract that calls it a "work made for hire" has no legal effect.

The Assignment Fallback (And Why the Clause Above Uses It). The clause quoted above includes a crucial fallback: "To the extent any Work Product does not qualify as a work made for hire under applicable law, Contractor hereby assigns to Company all right, title, and interest therein." This dual structure — first claiming work-for-hire, then assigning if that fails — is standard in professionally drafted contracts precisely because it provides a backup when the work-for-hire label does not stick. For independent contractors, the work-for-hire label often does not stick (because most deliverables are not in the 9 categories), but the assignment clause does the same practical work of transferring ownership.

Tax and Benefits Implications. Courts and the IRS look at whether a hiring party treats a worker as an employee (withholding taxes, providing benefits, issuing W-2s) vs. independent contractor (issuing 1099s). Calling a worker an "independent contractor" in a services agreement does not make it legally true. If the work arrangement has the characteristics of employment — the hiring party controls how and when the work is done, provides equipment, limits the worker from other engagements — misclassification exposure under federal and state law is significant. California's AB5 test (the "ABC test") is particularly strict: a worker is an employee unless (A) they are free from control, (B) the work is outside the company's usual business, and (C) they are customarily engaged in an independent trade. Under AB5, most creative freelancers working for a tech company are employees, which means the work-for-hire doctrine's employee path — not the contractor path — applies.

An IP assignment clause is the primary mechanism by which one party transfers intellectual property ownership to another. Unlike a license — which grants usage rights while the creator retains title — an assignment permanently transfers ownership. Understanding what is being assigned, when the assignment occurs, and whether it is valid are the key analytical questions.

Present Assignment vs. Agreement to Assign. Courts distinguish sharply between a present assignment of existing or future IP ("Consultant hereby assigns") and an agreement to assign future IP ("Consultant agrees to assign"). A present assignment of future IP is effective immediately when the IP comes into existence, without any further act by the assignor — this is the majority rule under cases like Speedplay, Inc. v. Bebop, Inc., 211 F.3d 1245 (Fed. Cir. 2000). An agreement to assign requires a further affirmative act (execution of a separate document) before title transfers, which creates a window during which the assignor could convey the IP to a different party. The clause above uses present-tense "hereby assigns" — this is a present assignment that automatically applies to Work Product as it is created.

Scope of Assignment — Worldwide, All Media, All Rights. Well-drafted assignments cover: all jurisdictions worldwide (not just the U.S.); all current and future forms of media and technology; all existing IP rights and rights that come into existence in the future; all derivative works and improvements. A narrowly drafted assignment (e.g., "for use in North America") leaves the assignor owning the IP everywhere else, which creates problems when the business wants to expand globally.

Consideration Requirements. Under general contract law, an assignment must be supported by consideration to be enforceable. When the assignment is part of an initial services agreement, the services fee provides the consideration. When an employer asks a current employee to sign an IP assignment agreement — after the employment has already begun — courts require independent consideration beyond continued employment in some states. California, for example, requires additional consideration (a bonus, raise, or other benefit) if an IP assignment is added to an existing employment relationship mid-stream. Signing an IP assignment at the start of employment, before any work begins, is supported by the employment itself as consideration.

Scope Limitation — What You Are Assigning. The most critical negotiation point for freelancers and consultants is scope: does the assignment cover only work created specifically for this client ("Project IP"), or does it sweep in everything the contractor creates during the engagement period regardless of connection to the client's project ("moonlighting assignment")? Some corporate agreements include language assigning all IP created during the engagement — even on personal time, using personal equipment, for completely unrelated purposes. This is often unenforceable under state law (see California Labor Code § 2870), but it creates litigation risk and chilling effects on independent creative work.

Moral Rights (17 U.S.C. § 106A — VARA; International). Moral rights are the right of an author to be credited for their work (attribution) and to object to distortion, mutilation, or modification of the work that would harm their honor or reputation (integrity). In the United States, moral rights under the Visual Artists Rights Act (VARA) apply only to works of visual art in a single copy or limited edition of 200 or fewer signed copies — not to works made for hire, commercial design, software, or most common contracting deliverables. Internationally, moral rights are far broader and often inalienable — EU, UK, and Canadian law protect moral rights for authors of most copyrightable works, and the rights cannot always be waived by contract. If you are contracting internationally or creating visual art, moral rights provisions in the contract require careful review.

Moral Rights Waiver in Contracts. Even though VARA moral rights are narrow in the U.S., many contracts include a broad waiver: "Contractor hereby waives any and all moral rights with respect to the Work Product." For purely commercial work within the U.S., this waiver is largely academic — the rights being waived usually do not apply to begin with. For internationally distributed creative work or fine art, the waiver has real practical effect. A contractor creating visual art for an international campaign should understand what moral rights they are waiving and whether any of those rights are inalienable under applicable foreign law.

A license grants usage rights to intellectual property while the licensor retains ownership. Licenses can be structured with extraordinary flexibility — exclusive or non-exclusive, for specific fields of use, limited to defined territories, for specified durations, with or without sublicensing rights, and revocable or irrevocable. Each dimension of the license structure has commercial and legal consequences that must be understood before signing.

Exclusive vs. Non-Exclusive. An exclusive license means the licensor grants rights to a single licensee and agrees not to grant the same rights to anyone else (including, if broadly exclusive, the licensor itself). A non-exclusive license means the licensor retains the right to grant the same rights to additional licensees simultaneously. Exclusive licenses in certain contexts can be registered with the USPTO and are treated similarly to assignments — an exclusive licensee has standing to sue for patent infringement without joining the patent owner in some circumstances. The distinction matters enormously in competitive markets: a non-exclusive license to use a competitor's technology gives you no advantage over that competitor.

Sublicensing Rights. The right to sublicense — to grant the licensed IP rights to third parties — is separate from the primary license and must be expressly granted. Without a sublicensing right, a licensee cannot authorize contractors, affiliates, or subsidiaries to use the licensed IP. For software licenses, this means a company cannot allow its contractors to access licensed tools without an express sublicensing right. The clause above is non-sublicensable — the licensee cannot extend the license to any third party.

Field-of-Use Restrictions. A field-of-use limitation restricts the license to a defined commercial application. For example, a pharmaceutical company might license a drug compound patent "for use in treating cardiovascular disease only" — if the licensee develops applications for oncology using the same compound, it needs a separate license for that field. Field-of-use restrictions are particularly common in technology licensing and pharmaceutical agreements. For the licensee, exceeding the field of use is infringement even if the licensee holds a valid license for other uses.

Territory. IP licenses can be geographically restricted — "worldwide," "North America only," "European Union," or any other defined geographic area. Since IP rights are national (a U.S. patent protects only within the U.S.; a European patent protects only in designated European states), a worldwide license technically requires separate rights in each jurisdiction, though this is usually handled contractually by the licensor representing that it has the right to grant worldwide use.

Royalties and Payment Terms. License agreements can involve upfront lump-sum payments, ongoing royalties (typically a percentage of net sales of products incorporating the licensed IP), milestone payments tied to development or commercialization events, minimum annual royalties, and combination structures. Royalty rates in technology licensing typically range from 1% to 10% of net sales depending on the IP's importance to the product, the exclusivity of the license, and the negotiating leverage of the parties. Patent licenses for foundational technology in mature markets often carry higher rates.

Term and Termination. A license "for the term of this Agreement" terminates when the underlying agreement expires or is terminated. A license "for the life of the patent" survives termination of the services agreement. Licenses linked to the underlying agreement expire when the relationship ends — which can be catastrophic if the licensee has built products or services dependent on the licensed IP. For any critical technology license, negotiate for an irrevocable license that survives agreement termination (or at minimum, a wind-down period sufficient to find alternatives).

License vs. Assignment: When One Matters More. From the licensor's perspective, a license is preferable to an assignment because the licensor retains ownership and can potentially recapture the rights. From the licensee's perspective, an assignment is preferable to a license because ownership cannot be revoked. The choice is also relevant for bankruptcy: in a bankruptcy proceeding, the trustee can reject (terminate) executory contracts — which can include licenses — under 11 U.S.C. § 365. Section 365(n) protects licensees of intellectual property (with some important exceptions and limitations) by allowing them to retain their rights notwithstanding the licensor's bankruptcy rejection. Understanding this protection — and its limits — is important in any significant technology license.

Background IP — also called "pre-existing IP" or "retained IP" — is intellectual property that the contractor or licensor owned before the engagement began or developed independently outside the scope of the current project. The failure to define and protect background IP is one of the most common and costly mistakes freelancers and consultants make in contract negotiations.

Why Background IP Carve-Outs Are Non-Negotiable. If a software developer signs a broad IP assignment with no background IP carve-out, and uses their own previously developed code libraries, frameworks, design systems, or tools as part of the client deliverable, those background IP assets are potentially assigned to the client along with the new work. The developer may lose ownership of tools they developed independently over years — tools they rely on for multiple client engagements. The same risk applies to graphic designers with pre-existing illustration elements, copywriters with pre-existing content frameworks, consultants with pre-existing methodologies, and any contractor who builds on existing IP to create new deliverables.

Defining Background IP Precisely. A well-drafted background IP clause should define background IP by reference to: (1) a schedule or exhibit listing specific pre-existing assets (preferred — provides certainty); (2) a general description of the types of pre-existing assets (e.g., "code libraries developed prior to [date]"); (3) a time boundary (created before the commencement date of the engagement). The more specific the definition, the less room for dispute. A blanket carve-out ("all IP owned by Contractor before this Agreement") is better than no carve-out, but a schedule of specific assets is best.

The License-Back Trap. Even with a background IP carve-out, the contractor must be careful about what license they grant back to the client for background IP incorporated in the deliverables. The clause above grants a "non-exclusive, royalty-free, perpetual, irrevocable license to use such Background IP solely as incorporated in the Deliverables." This is reasonable — the client needs to be able to use the finished product, and the license is limited to the deliverables as delivered. But some contracts try to expand this to: "a non-exclusive, royalty-free, perpetual, irrevocable license to use, modify, and create derivative works of the Background IP for any purpose" — which effectively destroys the value of the carve-out by giving the client a broad standalone right to the contractor's tools.

Improvements and Derivatives. Who owns improvements to background IP made during the engagement? If a contractor uses their existing code library and makes improvements to it while building the client's product, those improvements may fall under the IP assignment clause if they were made "in connection with the Services." A careful contractor will negotiate that: (1) improvements to background IP belong to the contractor; (2) the client receives a license to the improved background IP as incorporated in the deliverables, but not a standalone right to the improvements; (3) improvements developed specifically at the client's direction using client resources are an exception and belong to the client.

Improvements to Client's Background IP. The flip side: if the client provides its own existing codebase or IP assets for the contractor to work with, any improvements to the client's IP — enhancements, extensions, bug fixes — typically belong to the client, not the contractor. A properly structured agreement addresses both directions: the contractor's background IP is carved out and licensed back; the client's background IP remains owned by the client and improvements thereto are assigned to the client.

Portfolio Rights. Many creatives need to show their work to potential clients — designers show portfolios, writers show clips, developers show code examples. If all work product is assigned to the client and the contract does not include a portfolio license, the contractor technically cannot display or reference the work without the client's permission. Negotiate an explicit portfolio license: "Contractor retains the right to display and reference the Deliverables in Contractor's portfolio for the purpose of soliciting future business, subject to any confidentiality obligations that apply to the content of the Deliverables."

Joint IP ownership arises when two or more parties contribute to the creation of a work. It is one of the most practically problematic IP arrangements in commercial contracts — primarily because default legal rules governing joint ownership in the United States are markedly different from international norms, and because those default rules often produce outcomes neither party anticipated or intended.

U.S. Default Rules for Joint Copyright Ownership. Under 17 U.S.C. § 101 and § 201, a "joint work" is a work prepared by two or more authors with the intention that their contributions be merged into inseparable or interdependent parts of a unitary whole. Each co-owner of a joint copyright has an independent right to exploit the work — including license it to third parties — without the consent of the other co-owner, subject only to a duty to account to the other co-owner for profits. This means: if you co-own a software copyright with a business partner and the relationship sours, your partner can license the code to your competitor, without your consent, as long as they give you half the license fee. This is often a shocking result for parties who assumed joint ownership meant joint decision-making.

U.S. Default Rules for Joint Patent Ownership. Under 35 U.S.C. § 262, each co-owner of a patent may make, use, offer to sell, or sell the patented invention without the consent of, and without accounting to, the other co-owner. Unlike copyright co-ownership, patent co-ownership does not even require profit-sharing. A co-owner of a patent can therefore license it to competitors for free, manufacture infringing products, or simply do nothing — all without the other co-owner's consent or any obligation to share revenue.

International Joint Ownership Rules. Most other major jurisdictions are more restrictive: in the EU and UK, copyright co-owners must jointly consent to most exploitation; in France, joint owners must consent to any commercial exploitation beyond the co-owners' direct use. A joint ownership arrangement that seems workable under U.S. law may be administratively unworkable under international IP law if the jointly owned assets are commercially exploited globally.

The "I'll Own It and License It Back to You" Alternative. In practice, the cleanest structure for jointly developed IP is usually to allocate sole ownership to one party and grant the other party a license. The choice of which party owns depends on commercial realities: if one party contributed the foundational IP and the other contributed refinements, the foundational contributor may take ownership and license to the other. If the IP is equally important to both parties' businesses, a joint venture entity can be created to own the IP, with both parties licensing from the entity. Both alternatives are cleaner than default joint ownership because they avoid the "each party can do whatever they want" problem.

Allocation of Jointly Developed IP in Contracts. When joint ownership is unavoidable or desired, the contract should override the default rules by specifying: (1) Neither party may license, assign, or transfer their interest without the other party's written consent; (2) Any licensing revenue is shared equally (or per a specified ratio); (3) Any decision to enforce the IP against third-party infringers requires mutual agreement; (4) Exit mechanisms — what happens to the jointly owned IP if one party exits the relationship, wants to acquire the other's interest, or goes bankrupt.

The "Necessary Contributions" Test for Copyright Joint Authorship. An author's contribution must be copyrightable to give rise to joint authorship — an idea, concept, or suggestion without copyrightable expression does not create joint ownership rights. This matters in creative collaborations: a client who provides detailed written specifications, wireframes, and design direction that the contractor implements may or may not be a joint author depending on whether the client's contributions crossed the threshold of copyrightable expression vs. unprotectable ideas. Courts have split on this, but the majority position (following Childress v. Taylor, 945 F.2d 500 (2d Cir. 1991)) requires that each co-author intended the work to be jointly authored at the time of creation.

Open source software (OSS) is incorporated in virtually every software project. Understanding the difference between copyleft and permissive licenses — and the contractual risks associated with each — is essential in any software development, consulting, or technology licensing agreement.

Permissive Licenses (MIT, BSD, Apache 2.0). Permissive licenses allow the licensee to use, modify, and distribute the licensed software — including in proprietary, commercial products — with minimal conditions. Typical conditions: attribution (including the original copyright notice), disclaimer of warranties. A product incorporating MIT-licensed code can be distributed as closed-source proprietary software without triggering any source code disclosure obligation. The Apache License 2.0 adds patent grant provisions: contributors provide an express patent license for contributions they make, which can be important in patent-heavy technology areas.

Weak Copyleft Licenses (LGPL, MPL). The GNU Lesser General Public License (LGPL) and Mozilla Public License (MPL) impose copyleft obligations on modifications to the licensed components themselves, but generally permit incorporation in a larger proprietary product without "contaminating" the proprietary code. The LGPL requires that users be able to modify and relink the LGPL-covered library — relevant for statically linked binaries, but generally not for dynamically linked libraries. MPL requires any changes to MPL-covered files to be made available under the MPL.

Strong Copyleft: GPL and AGPL — The Contamination Risk. The GNU General Public License (GPL, versions 2 and 3) imposes a powerful "viral" obligation: if you distribute a product that incorporates GPL-licensed code (or a derivative work of GPL-licensed code), you must distribute the source code of the entire product — including your proprietary code — under the GPL. This is the "copyleft" or "viral" effect. The GPL does not prohibit commercial use, but it prohibits proprietary distribution of works that incorporate GPL code. The Affero GPL (AGPL, version 3) extends this obligation to software used over a network: if users interact with AGPL-covered software over a web service, the source code must be made available even if the software is never "distributed" to users. This matters enormously for SaaS products.

Practical Business Impact. If a contractor incorporates GPL-licensed code into client software, the client's entire product may become obligated to be distributed under the GPL — meaning the client must release its source code and cannot maintain a proprietary closed-source product. The warranty clause above directly addresses this risk: the contractor is representing that the deliverables contain no open source that would trigger these obligations. Breach of this warranty — even inadvertent inclusion of a GPL component — can be a material breach of contract.

OSS License Compatibility. Open source licenses interact with each other: GPL-licensed code cannot be combined with code licensed under terms incompatible with the GPL. Combining MIT and Apache-licensed code with GPL code in the same binary is generally possible; combining GPL and GPL-incompatible licenses creates insoluble legal problems. AGPL and GPL have compatibility issues with certain other copyleft licenses. Development teams combining multiple open source components should use an OSS license compatibility matrix to identify conflicts.

Dependency Management and Audit. Modern software products incorporate dozens or hundreds of open source dependencies through package managers (npm, Maven, pip, Go modules). A single transitive dependency — a dependency of a dependency — that is GPL-licensed can create contamination risk if it is statically linked or otherwise combined with the product code. Production contracts should include an obligation to conduct an OSS license audit (using tools like FOSSA, Black Duck, or Snyk License) before delivery, and should address remediation obligations if GPL-contaminated code is discovered. The contractor should represent and warrant the results of the audit.

Trade secrets are the most broadly applicable category of IP protection — virtually every business has trade secrets, even if it holds no patents or registrations. Understanding the contractual obligations that flow from trade secret law, and how they interact with NDA and confidentiality provisions, is essential in any commercial relationship.

The Defend Trade Secrets Act (DTSA), 18 U.S.C. § 1836. The DTSA, enacted in 2016, provides a federal civil cause of action for trade secret misappropriation. Key elements: (1) the information must qualify as a "trade secret" — meaning it derives independent economic value from not being generally known or readily ascertainable, and is subject to reasonable measures to maintain secrecy; (2) the information must be misappropriated — acquired by improper means (theft, bribery, misrepresentation, breach of duty, espionage) or disclosed/used without authorization while knowing or having reason to know of the improper acquisition. DTSA remedies include: actual damages, unjust enrichment, royalty-based damages, exemplary damages up to 2x actual damages for willful/malicious misappropriation, and attorney's fees for willful misappropriation or bad-faith claims.

State Trade Secret Law: The Uniform Trade Secrets Act. Virtually all states have enacted some version of the Uniform Trade Secrets Act (UTSA), though with variations. The DTSA does not preempt state law, so both federal and state claims can be pursued simultaneously. Key state variations: California's UTSA (Cal. Civ. Code §§ 3426–3426.11) broadly defines trade secrets and is frequently litigated; Texas and Florida also have robust state trade secret remedies. Practitioners should understand both federal DTSA and applicable state law, as they may provide different remedies, statute of limitations periods (DTSA: 3 years; state law varies), and procedural advantages.

"Reasonable Measures" — The Critical Qualification. A crucial element of trade secret protection is that the owner must take "reasonable measures" to maintain secrecy. Contractual confidentiality obligations contribute to this requirement — having employees and contractors sign NDAs is a standard "reasonable measure." But reasonable measures go beyond just having NDAs: physical security, access controls, data encryption, password protection, need-to-know access restrictions, and regular training on confidentiality obligations are all part of a reasonable measures program. A company that shares its "trade secrets" with dozens of parties under informal verbal agreements, stores them on publicly accessible servers, and fails to label confidential documents may find that a court determines the information was not actually maintained as a trade secret.

NDA Integration. Most NDAs reference trade secrets explicitly because: (1) trade secret protection requires active steps to maintain secrecy (a signed NDA documents one such step); (2) breach of the NDA's confidentiality obligations and DTSA misappropriation are related but distinct legal claims — a party can pursue both; (3) the NDA typically provides a contractual basis for injunctive relief, attorney's fees, and liquidated damages in addition to DTSA remedies. The clause above cross-references the DTSA explicitly to ensure trade secrets are clearly within the scope of the confidentiality obligations and to reinforce the availability of federal remedies.

DTSA Immunity Notice Requirement. One often-overlooked DTSA provision: under 18 U.S.C. § 1833(b)(1), an employee (or contractor acting as an employee) cannot be held criminally or civilly liable for disclosing trade secrets in confidence to a government official or attorney for purposes of reporting a suspected violation of law. Employers who want access to DTSA exemplary damages and attorney's fees in litigation must provide notice of this immunity in any NDA or agreement that governs use of trade secrets. The notice can be included directly in the NDA or by cross-reference to a policy that includes the required language. Many template NDAs omit this notice, creating a gap in the employer's DTSA remedies.

Inevitable Disclosure Doctrine. Some courts — primarily in Illinois and a minority of other states — apply the "inevitable disclosure" doctrine to trade secrets: a former employee can be enjoined from working for a competitor even without proving actual misappropriation, if it is inevitable that they will use or disclose the former employer's trade secrets in their new role. California expressly rejects the inevitable disclosure doctrine. Understanding whether your state applies this doctrine — and including appropriate language in employment agreements — is relevant for both employers seeking to protect trade secrets and employees seeking to change jobs.

IP warranties and representations are the contractor's or licensor's formal assurances about the quality, ownership, and non-infringement status of the IP being delivered or licensed. Breach of an IP warranty can expose the warranting party to substantial liability even if the breach is entirely innocent — which is why understanding the scope of IP warranties before accepting them is critical.

Non-Infringement Warranty. The most consequential IP warranty is the non-infringement representation: that the IP does not infringe any third party's intellectual property rights. This warranty is typically absolute — not qualified by "to Contractor's knowledge" or "as far as Contractor is aware." An absolute non-infringement warranty means the contractor is strictly liable if infringement occurs, even if the contractor had no knowledge of the conflicting IP and conducted a thorough clearance search. In the patent context, this is particularly significant: a contractor creating a new software product has no practical way to guarantee it does not infringe any of the hundreds of thousands of software-related patents in the USPTO database, many of which have broad and unclear scope.

Originality Warranty. The representation that deliverables are "original works of authorship" means the contractor has not copied from any third-party source — not from other clients' code, not from proprietary databases, not from online code snippets with incompatible licenses. In the era of AI-assisted content generation, this warranty creates additional complexity: if a contractor uses an AI tool to generate portions of the deliverable, the output may: (1) reproduce third-party copyrighted training data; (2) be uncopyrightable because it lacks human authorship under current U.S. Copyright Office guidance; (3) trigger specific representations about AI-generated content. Clients should specifically ask contractors whether AI tools were used in creating deliverables, and contracts should address AI-generated content directly.

Authority to Grant Warranty. The contractor must warrant that they actually have the right to make the assignments and grant the licenses promised in the contract. This can be breached if: (1) the contractor has a prior agreement with another client that already assigned the same IP; (2) the contractor used a subcontractor who retained ownership; (3) the contractor's background IP is jointly owned with a third party who has not consented to the grant; (4) the contractor has a prior employer agreement that already assigned the IP. Prior agreements that limit a contractor's freedom to assign new IP — common in employment agreements — are a frequent source of breach of authority-to-grant warranties.

Knowledge-Qualified vs. Absolute Warranties. Absolute warranties ("Deliverables do not infringe any third party's rights") impose strict liability — breach occurs regardless of the warrantor's knowledge or care. Knowledge-qualified warranties ("to Contractor's knowledge, Deliverables do not infringe") require that the warrantor actually knew about the infringement to be liable. For contractors, knowledge-qualified warranties are significantly less risky, particularly for non-infringement (where exhaustive clearance searches are often impractical). Negotiate for knowledge-qualified representations wherever possible, particularly on non-infringement.

The Warranty Disclaimer and Its Limits. Many contracts include a warranty disclaimer: "EXCEPT AS EXPRESSLY SET FORTH HEREIN, THE DELIVERABLES ARE PROVIDED 'AS IS' WITHOUT WARRANTY OF ANY KIND, AND CONTRACTOR DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE." This language effectively limits the contractor's warranty obligations to those expressly enumerated in the agreement. Warranty disclaimers must be "conspicuous" under the UCC (typically in all caps) to be enforceable against a merchant. Note that warranty disclaimers do not negate the express representations in the same contract — they disclaim implied warranties, not explicitly stated ones.

IP indemnification is the contractual mechanism for allocating who pays — and who defends — when a third party asserts an infringement claim against one of the contracting parties. It is the financial consequence provision that backs up the IP warranties, and it is frequently where the real commercial risk lies.

The Three Elements of an IP Indemnity. A complete IP indemnification obligation includes: (1) the duty to indemnify — to compensate the indemnitee for losses, damages, and expenses incurred; (2) the duty to defend — to take over the defense of any covered claim (hire lawyers, manage the litigation) at the indemnitor's expense; (3) the duty to hold harmless — to prevent the indemnitee from incurring covered losses in the first place (effectively a subset of indemnification). All three may or may not be included in a given clause. A duty-to-defend is particularly significant because it shifts litigation management — not just financial exposure — to the indemnitor.

Mutual vs. One-Way Indemnification. IP indemnification in technology contracts may be mutual (each party indemnifies the other for its own IP) or one-way (the contractor indemnifies the client for infringement in the deliverables). One-way indemnification — the contractor indemnifying the client — is standard. Mutual IP indemnification becomes relevant when the client is providing its own IP (existing platform code, datasets, brand assets) that the contractor is incorporating, creating a risk that the client's provided materials infringe third-party rights.

The Cap on IP Indemnification Liability. The clause above caps the contractor's IP indemnification liability at the total fees paid under the agreement. This is often a heavily negotiated provision. Clients typically want uncapped IP indemnification — the contractor is solely responsible for its creative output and should bear the full cost of any infringement claim. Contractors counter that they cannot indemnify a $5 million company on a $10,000 engagement — the asymmetry makes uncapped indemnification commercially unreasonable. Common compromise positions: cap at a multiple of fees paid (2x or 3x); cap at the contractor's professional liability insurance coverage; carve out willful infringement from the cap.

Carve-Outs from IP Indemnification. Any well-drafted IP indemnification from a contractor should include carve-outs for infringement caused by: (1) modifications to the deliverables made by the client or its contractors after delivery; (2) combination of the deliverables with client-provided materials, third-party software, or hardware not provided by the contractor; (3) continued use of the deliverables after the contractor has provided a non-infringing replacement or modification; (4) use of the deliverables in a manner or for a purpose not specified in the agreement. These carve-outs track the standard "sole basis" infringement scenarios where the contractor's deliverables would not infringe in isolation — only the client's actions create the infringement.

Notification and Control Obligations. IP indemnification clauses typically impose procedural requirements on the indemnitee: prompt written notice of the claim; cooperation with the indemnitor's defense; and the indemnitor's right to control the defense and settlement of covered claims. Failure to provide prompt notice — allowing the claim to develop without the indemnitor's knowledge — can impair or waive the indemnity in some jurisdictions if the delay prejudiced the indemnitor's ability to defend. "Control of defense" provisions mean the contractor (not the client) gets to choose the defense lawyers and negotiate settlement — important because the contractor may have interests that conflict with the client's interests in how a settlement is structured.

IP Insurance (Errors and Omissions / Professional Liability). For contractors accepting IP indemnification obligations, professional liability insurance (also called errors and omissions or E&O insurance) is the primary risk management tool. E&O policies typically cover claims arising from professional services, including copyright infringement and IP-related errors. Review the policy's IP coverage carefully: some policies exclude intentional infringement, patent infringement, or certain categories of IP claims. Contractors accepting significant IP indemnification obligations should maintain E&O coverage adequate to satisfy those obligations.